Several studies indicate that women with breast cancer who undergo radiotherapy are susceptible to secondary lung cancer, whether they are smokers or nonsmokers. However, all studies to date have methodological limitations and have been small. Also, none have used molecular markers, which can improve exposure assessments or elucidate mediating mechanisms. Over time, radiotherapy methods have changed and doses to the lung have lessened. On the other hand, prevalence of smoking has increased among women in the western world. The identification of lung cancer risk is important in the context of the debates for benefits of radiation therapy in good prognosis tumors or older women. Thus, a study of breast and secondary lung cancer is needed to improve dosimetry assessments for radiation induced lung cancer, with and without an interactive effect of smoking. Also, studying a unique population of women who have had both breast and lung cancer can provide new insights into carcinogeneis and cancer risk. In order to do this, we are proposing a population-based study using the Swedish Cancer Registry (SCR) and determination of radiation doses to the whole lung and side of the lung where the tumor subsequently develops. Reliable smoking data will be available. Our specific aims are to: 1) determine risk factors for secondary lung cancer in women treated with radiotherapy for breast cancer using complementary nested case-control and case-only study designs (n=559 cases and 559 matched controls); 2) to determine p53 inactivation pathways, (i.e., mutational spectra and loss of heterozygosity) in lung tumors of women with a prior history of breast cancer (n=402) and; 3) to determine the frequency of p53 inactivation pathways in breast tumors of women who did and did not develop lung cancer, and compare them to the frequency of p53 inactivation pathways in the lung tumors (n=342 cases and 342 controls). The first aim will allow us to identify risks. The second aim will provide information about the mechanistic relationship of radiotherapy to lung cancer and may identify a unique spectrum for radiation-related lung cancer. The third aim considers the combined occurrence of breast and lung cancer in a woman as phenotype of susceptibility for multiple primary cases. This study provides unique opportunities. Using the SCR and the unparalleled ability to obtain tissue blocks dating back to the 1950's, we can provide new data to understand risk in the context of molecular markers, especially because we will be able to retrieve the tumor blocks from both the breast and lung cancer from the same women.